Document 9823

CEREBRAL SINUS THROMBOSIS IN
CHILDREN
BY
J. H. EBBS, M.D., D.C.H.,
The Children's Hospital, Birmingham.
Thrombosis of the cerebral venous sinuses has been recognized for many
years. The first description is credited to Morgagnil in 1717, and
Abercrombie' made the first accurate study of the condition. It is usually
classified as primary or secondary. The primary type is described as
occurring with marasmus, dehydration, gastro-enteritis and anaemia;
whereas the secondary type is always associated with infection, chiefly of
the middle ear and mastoid region, and although it is a not uncommon
sequel of infection spreading from the face, scalp, skull bones or meninges,
it rarely follows trauma to the skull. Recently Ehlers and Courville' have
described a combined type which is found in marantic, dehydrated infants
with otitis media. Holt and McIntosh4 state that, 'it is customary to
distinguish between inflammatory or septic thrombosis and a so-called
marantic or cachetic type. In the former, obvious inflammatory changes
are found in the thrombus; in the latter, evidences of inflammation are
inconspicuous. Infection in some part of the body is always present, however, and even here it is probably of etiological significance.'
Byers and Hass' give a complete review of the subject while reporting
fifty cases which had been observed in Boston during a period of six years.
They found twenty-six cases associated with infection and twenty-four which
they classified as primary, one half of the latter occurring in August,
September and October. Sixteen of the twenty-four cases showed a severe
unexplained diarrhoea. Acute dehydration was a more important factor
in these cases than malnutrition. Wyllie' has described three cases at the
Hospital for Sick Children, Great Ormond Street, two of which were
associated with pneumonia and one with infantile diarrhoea. He classifies
the origin of the disease as otitic and marantic. The latter group includes
those cases secondary to infantile diarrhoea, prolonged suppuration,
tuberculosis, carcinoma, pertussis, chlorosis, empyema, pneumonia,
appendicitis, cerebral tumour, infectious diseases and following operations.
Ehlers and Courville' have reviewed the literature and have collected sixty
cases of primary thrombosis of the internal cerebral veins. In the majority
the thrombosis of the internal cerebral vessels was associated with a similar
condition of the cerebral venous sinuses. Twenty-one of these cases
occurred in infants or children. Simpson7 analyzed forty-two cases of
thrombosis of the cerebral vessels which had been found at the Hospital
for Sick Children, Toronto, during a period of ten years. Some text-books
of diseases of children do not mention the condition, whilst others refer
only to thrombosis associated with mastoid infection. Some state that
symptoms may be entirely lacking, the condition being found post mortem
134
ARCHIVES OF DISEASE IN CHILDHOOD
in association with some other process. The usual picture described, however, is one of marked septic symptoms, i.e., chills, high temperature,
headache and meningeal signs. Osler8 described the syndrome of mental
dullness, headache, convulsions, vomiting and paralysis. If the superior
loingitudinal sinus is affected paralysis of the legs may be the diagnostic
point since the cortical supply to the legs is close to the superior longitudinal
sinus. Recovery rarely occurs under these conditions and when it does a
permanent spastic paralysis of the legs usually results.
Thrombosis of the cerebral vessels in children is not usually considered
as a common occurrence, but during the past two years fourteen patients
have been admitted to the Birmingham Children's Hospital suffering from
this condition. The records of the hospital for the previous twelve years
show that only eighteen cases had been found. It would therefore appear
that there has been an increase in the incidence of the disease, but it still
remains as an entity which is seldom considered in the differential diagnosis
of cerebral conditions in young children. It is important if possible to
diagnose the condition during life, and a review of several recent cases brings
out many points which have been found of some value in achieving this end.
Case records
A summary of the age, etiology and important clinical and pathological
findings has been made of the thirty-two cases recorded in this hospital
during a period of fourteen years (table 1). The case records of eight
patients are appended, and the fourteen cases observed in the past two
years form the basis of the remarks which follow.
Case 19. M. O., a girl, aged two years, was well until two days before
admission, when she became restless and tired. Her breathing became rapid
FIG. 1. Case 19. M. O.-Thrombosis of the internal cerebral vessels
resulting in hemiplegia,
~e
CEREBRAL SINUS THROMBOSIS IN CHILDREN
4)C
r
C2A
.~ ~ ~ '
CD
4)~~~~~~~~~~-
4A
a)
~~~4)~
Ca
Ca
'4
...
=
CsP=
0
4)'
=544,S
) ;;
E
0_
~
-*r
4)
O
on
4._
ai~~~~~~4
'C
_a
,_
A
05.3
r
._)
4)
X
rn
Co
a)
._
(i)
a3
-
*
-;
O,!
4)
0)
ce~~~~~~~~~~4
db
c5
S4
ce
12)
-
Co)
Q
;4
94)
c
C)~~
~
L"
0.
n.
)c3
rn~
~
~
r
V
;AN~~~~4
S
Cs~~~Or
0
ro
*
i
,3.
0O
O
Co
.4
ce
-.
04
'4
Cn
°4.
Cs
4a,
to
-4
-14
r
c
O-.
co
Cn
4)
--4
4;
4)
-
> s~~0 =
i
@
;
._
X 4)
0,
4) 4)
4)
:
n
._
)
*4)
w--
4
rm
Cd
0
U
~
-0
Sa
'-4''
,-
'4
4
.~~~~
-c
-S
4)
1-
C)
)3
'44. y
'4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1
So~~~~~~~~~~~2~
0
.4
'4
'4 4)
)
;L4
clito~)4.~ ~ ~ -
'4:
i,
4)
135
-4)
Cf2
t-
q.
.
c
a) S-
E
5C
P4
CD
).:
i
A2
136
ARCHIVES OF DISEASE IN CHLDHOOD
ku
Ca
'4
0
.14
4)
'4
0
.00
~.%~
Cd
m
o C4)
k
._
0 as
*
Ca
4).0
b6,o~
4)9
*-
o
;C)
8~-
0"o
04
0
Ca
z
-
.r04=
4)~i
'4
k)
so4)
.0
4,
0~~~~~
4)
d
O
d
-°
E
04~~~~
.-
o
1
0~~~~
>.
Ca
4,0
0
*n
0
m
~~~~~m
C4,
0
4)~~~~~~~~~~~~~~
,0
..
r
4,Ca &
on-0
0
I2
.a3
4
cm'4
Ez
0
t
0
o
4
0
E-L
C454a
~
0
'
Ca Ca
;-4
*tCI2
;
)
04Cs
4
4
cn).
-2
4a
-.
00
~~4
4a
= w C
-
;
Gil
4
hO
04 ~
~
~
4,
4
0
._
04,-40
-
4)~~~~~C
E
'4c
60.0~
4)
4,
*'4
..-
0
-Q-
0 60
4)4,4)
S
0
Caa
o'
t:0
0
04
'4
O
0,0
W
.04)5
a)4,
0
-
e'4
0
C)
0
Ca
._
*
0.4
.
0
.
00;
604
4
4
Es4
4,
0
4,
)
4)'
k
z
7.1
0
0
Cs1 9
-9
045
.4
C.
(D
V.-
r-4
_
Ca
4)
cj..~
CO
S.4
>.a
P4
N
45
.4,
0
*P
m
1-
e
P-4
mK Y
-4
p-
-r;~
<,0
_4
4)
Ca
0~~~~~~~~~4
-4a4)
0
Ca 4
00-
00
>
Ln0~~~~~.
CI~c,2 Q~Ca
C~I2 0f
~
~ 0a
4-D
4)
.o
4).- ~
'4
4
.:
0.
°
-*
00_.
Ca
PE r.
Ca
O
0
4,
Ca~~~~
4)
.:
4)
0
p
0
._
Ca
Cs Ca00
0
'4
0
.04
X
q
1-
~60
0~~~~~~~6
I
.
-;
F-f0
<
0)~
Om
C-9
-
9
4
CEIR-EBRAL SINUS TH4ROMBOSIS IN CIIIDR1tN
~
~~
~
e
ir
'4
~~~~~~~~~~~~~~~~~e
'4
~
~
Z4)
c
~
~
4) 0
4D)2 :0
a,5
-'~~~~~
~
c4~~~~~~~~~..a
Cac
-4
Ca0
4a~~~~~~~~~~~~'
44)'45
Ca
0~.
P4n
cr.
4--)
~
0~
0
S
~~~~~4)
0k m
4
k
--
4)4)
.)
c' -
4
)4
12
4)
'4
Ca
U
U)
.04)Cs
*4
O
rn es
4z~~~~~~~~~~~~~~~~'
0
Ca
Cdc aC
4
4A)
)
4a
AD w
. rn
t
V
o
t
4aV
_
CA
rr
0
0
mPv0
0
4) -=
~
40
'4
O
=bO
50;4
V
o
*o*
'4rvq
'4
'4.
4-4
0
4a
d
Ca
o
-Q
4)
So
(1v
4)
o
'4~~~~~~~~~4
'.-O
Cso
C
,=
0
;4)0
C3
05
~~~~0
= r
'4 ~~~~~~~~Z00
4)
bO
.
c3n
~._
*-
.C
0
40
C)~~'
0)
'.-
rn
c312
=0
*4C
9
~
~
~
~
~
~
~
~
V4)
4) go
0~
'4
44
4)
.!
;4)
4)
0C)
*0
Cs
=;{4
'4 4)
0
_
+0
*<e
0
an
ED
0
Ca
._40
0
S
Ul)
U7)
q
'4a'
0
S C
P
D0S
0>
-'4
1-1
04
O .4
U)
.01
_
+0
m
Is
I
P-
.1
2g~
V
-
~~~~~34~~~~~~~4
Ca
U)
I"u
Cs
0
1
.
tJ2
n
-
(12
~.
f
Y
_4N
.
-, .-4
Go
&P
Ca
.
ARCHIVES OF DISEASE IN CHILDHOOD
138
and she lapsed into unconsciousness in two or three hours. Twitching of the
arms and legs was then noticed. She regained consciousness in a short
time and vomited. Later in the day the child had a definite convulsion and
remained unconscious after the seizure. The morning after the onset the
right arm and leg were found to be weak. She was feverish from the onset
and the temperature remained about 1020 F. throughout. Convulsive
twitchings of the right side of the body continued after admission to the
hospital. A lumbar puncture revealed 1,000 creneated red blood cells per
c.mm. with no increase of white blood cells. Cultures were sterile. A blood
count showed 24,800 leucocytes with 81 per cent. polymorphonuclears and
evidences of slight toxic changes. Twitching of the right arm and leg
continued throughout the next four days. The fifth day after admission
she seemed improved and took feeds well, but the following day she became
drowsy and was very difficult to arouse. There was a slight stiffness of the
neck, stertorous breathing and an irregular pulse. Spasticity of all her
limbs developed. Her condition had rapidly become worse. A trephining
operation was done over the left meningeal area because cerebral
haemorrhage, tumour or abscess had to be considered in the differential
diagnosis of a child who was becoming rapidly worse. At operation nothing
was found beyond congestion and increased pressure. The following day
the child could not swallow, there was marked general rigidity, and she
died eleven days after the onset. A summary of the cerebro-spinal fluid
findings is shown in the following table.
TABLE 2.
CEREBRO-SPINAL FLUID FINDINGS IN CASE 19.
I)AAY
LE;UCOCYTES
per c.inmn.
1
2
5
6
18
5
8
8
RED BLOOD CEL,LS
per c.mnm.
per cent.
CHLORIDES
per ceint.
PROTEIN
per cent.
1,000
56 mgm.
876 mgm.
90 mgm.
-
9132 iiigmn.
125 mngm.
SUGAR
1,500
Slightly
xanthochromic
,,
_
Post-mortem examination revealed a terminal broncho-pneumonia of
the lower lobes of both lungs. The brain was markedly congested; all of the
superficial blood vessels were thrombosed, also the vessels in the substance
of the brain, chiefly on the left side in the region of the thalamus and internal
capsule. All of the venous sinuses were thrombosed (see figure 1). The
only focus of infection was a purulent exudate in the crypts of both tonsils
which were hypertrophied. Post-mortem blood cultures and cultures of the
thrombus in the superior longitudinal sinus remained sterile.
Case 20. J. B., a girl, aged sixteen months was admitted with a history
of having been well until the day before admission, when she appeared
flushed, became very restless and the following morning developed convulsions, which continued after admission to the hospital. The convulsive
twitchings were at first generalized, but later became confined to the left
side of the face, the left arm and leg. A lumbar puncture relieved the
convulsions. The cerebro-spinal fluid was clear, was under increased
pressure and contained 4 cells per c.mm., no sugar, chlorides 774 mgm. per
cent., protein 13 mgm. per cent. On examination there was no neck
rigidity, Kernig's sign was negative and the reflexes were normal. There
was no evidence of infection. A blood culture remained sterile. The child
continued to be very restless, resented being touched, and at times seemed
to be semi-conscious, although she could take fluids. The temperature was
normal and remained so until the day before her death, when it became
CERREBRAL SINUS THROMBOSIS IN CHILDREN
1839
elevated to 1020 F. The day after admission the left arm and leg became
paralyzed, the neck became stiff and the Kernig sign was positive on both
sides. The reflexes were absent on the left side. During the next four
days her condition became rapidly worse. In the late stages of her illness
bluish-black areas appeared on the neck, hands, ears and nose due to
thrombosis or emboli in the terminal vessels. A fine purpuric rash was
present over the lower legs and a few purpuric spots on the backs of the
hands.
The child died nine days after the onset. In view of the sudden onset
with convulsions and meningeal signs without cerebro-spinal fluid findings
of meningitis, and experience with the previous case, a diagnosis of cerebral
sinus thrombosis was made. At post-mortem examination the brain showed
a diffuse gelatinous haemorrhagic area in the region of both Rolandic areas
and extending over the cortex. This was more marked on the right side.
All of the superficial veins and the superior longitudinal sinus were
thrombosed (see figures 2 and 3). Purulent exudate was found in both
FIG. 2. Case 20. J. B.-Left cerebral hemisphere showing thrombosis
e.vinr Re; 1tAn
t1x7,1e
+1k assocnia£1ted subanrac{hnonid hae smorrhap- e
an
1'1G.
3. Uase 20. J. B.w-uperlor surface of right cerebral niemispnerei
140
ARCHIVES OF DISEASE IN CHILDHOOD
maxillary antra, ethmoid cells, middle ears and mastoid antra. This pus
was cultured and grew pneumococcus type 1, which was also cultured from
the clot in the superior longitudinal sinus.
The above two cases were outstanding in their sudden onset without
any evident infection during the course of illness. The cultures of
pneumococcus in one case and the presence of a septic tonsillitis in the other
are possible etiological factors.
Case 21. J. B., a girl, aged two years, was an only child of healthy
parents, who was well until the day before admission, when she became
feverish and was put to bed. At 3 a.m. she had a convulsion lasting about
ten minutes, with frothing at the mouth, eyes rolled up and twitching of
both arms and legs. At 6 a.m. she had a similar convulsion lasting twenty
minutes which was followed by vomiting of brown frothy fluid and upper
abdominal pain. She had not been in contact with any known infection.
When examined upon admission to the hospital the child was rolling her
head from side to side, but apart from this she showed no abnormal signs.
The temperature was 105 40 F., pulse 160 and respirations 48. The
temperature remained elevated between 1020 and 1030 F. for one week.
FIG. 4. Case 21. J. B.-Purpuric rash over right leg.
The only focus of infection was a mild pyelitis. The child vomited several
times daily for the first week but only occasionally subsequently. Four
days after admission the legs became hypertonic but there were no other
signs of disturbance of the nervous system. It was later noted that the
child appeared to be blind and she continually opened and closed her mouth.
This was associated with clonic spasms of both upper limbs, the right arm
aind hand being more affected than the left. The legs were normal but a
definite stiffness of the neck had developed. A lumbar puncture revealed
A9 red blood cells per c.mm. and 6 white blood cells per c.mm. The cerebrospinal fluid sugar was 48 mgm. per cent., the chlorides 766 mgm. per cent.
and the protein 23 mgm. per cent. The spinal fluid and urine were searched
for tubercee baccilli but none were found and the Mantoux test was negative.
The blood urea was 23 mgm. per cent. and the non-protein nitrogen 21 mgm.
per cent. A blood culture remained sterile during five days' incubation.
Cultures of the cerebro-spinal fluid were also sterile. The blood showed
90 per cent. of haemoglobin with normal colour index. The leucocytes
numbered 7,700 per c.mm. and the differential count showed 20 per cent. of
segmented polymorphonuclear cells and 40 per cent. non-segmented forms.
The W assermann reaction was negative. An x-ray examination of the
Jungs showed no evidence of pneumonia or tuberculosis. A purpuric rash
CEREBRAL SINUS THROMBOSIS IN CHILDREN
141
developed four days before her death which first appeared in both groins
and later over both legs (see figure 4). Twelve days after admission there
was a slight purulent discharge from the right ear which did not recur. The
legs gradually became more spastic and the head retracted. There were
frequent twitchings of both hands with an increased rigidity of the arms.
The left external jugular vein became very prominent and a thick firm
thrombus could be palpated throughout its entire length. The child
remained drowsy with long periods of semi-consciousness, a variable
temperature and gradual decline in general health until she died six weeks
after admission to the hospital.
At post-mortem examination the skin showed numerous purpuric spots
over the legs, arms and trunk, but most marked on the legs (see figure 4).
The lungs contained several embolic infarcts and haemorrhagic areas. Both
FIG. 5. Case 21. J. B.-Thrombosis of the superficial vessels and
subarachnoid haemorrhage.
kidneys showed septic embolic infarcts. All of the venous sinuses of the
skull were thrombosed. The thrombosis extended from the superior
longitudinal sinus along the lateral sinuses and down into the jugular vein.
The left external jugular vein was markedly distended and thrombosed
throughout its complete length. The straight sinus, inferior saggital and
veins of Galen were thrombosed. The majority of the large superficial veins
showed thrombosis. There was considerable subarachnoid haemorrhage
over the left occipital lobe, another over the right midparietal region and
a small one over the superior surface of the right frontal lobe (see figure 5).
The vessels in the left choroid plexus were completely thrombosed as were
the vessels in the surrounding brain substance. There was considerable
142
12ARCHIVES OF DISEASE IN CHILDHOOD
softening of the brain in this area (see figure 6). There was thick yellow
pus present in both middle ears and mastoid antra, from which a pure
culture of a haemolytic streptococcus was grown.
The outstanding feature of this case was an illness of six weeks' duration without any evidence of infection except pyelitis. Whereas at postmortem examination the mastoids were found to be grossly infected with
pus which grew a haemolytic streptococcus.
FiIG. 6. Case 21. J. B.-Thrombosis of choroid
plexus.
Case 22. S. W., a girl, aged two-and-a-half years, was well until four
days before admission, when she became tired in the afternoon and slept for
a long time. It was then noticed that she could not use her right hand
and could not speak. The following day the right hand and the right side
of the face showed marked twitching which continued until the day before
admission. The right leg was then noticed to be motionless. The child
was restless between long periods of drowsiness. In the hospital she
gradually became more comatose. The temperature ranged between 100°
and 1040 F. The right ear drum was pale but not bulging. There was
neck rigidity and a positive Kernig sign. The right side of the face showed
some weakness and the right arm was completely paralyzed and flaccid.
The right leg became spastic and when touched it would go into a clonic
spasm. The left side appeared normal until a few hours before death when
the left hand began to twitch. The cerebro-spinal fluid contained 8
leucocytes and 17 red blood cells per c.mm. The cultures were sterile.
Biochemical examination revealed sugar 70 mgm. per cent., chlorides 866
mgm. per cent. and protein 10 mgm. per cent. The plasma chlorides on the
same day were 695 mgm. per cent. A throat swab grew chiefly a haemolytic
streptococcus. A blood count showed haemoglobin 104 per cent.,
ervthrocytes 5,600,000 per c.mm., leucocytes 11,200 per c.mm. and the
differential count was as follows: Polymorphonuclears 58 per cent. (30 per
cent. non-segmented, 28 per cent. segmented), metamyelocytes 2 per cent.,
lvmphocytes 36 per cent. and monocytes 4 per cent. The urine examination showed the following: -albumin - cloud; sugar - none; acetone + +;
CEREBRAL SINUS THROMBOSIS IN CHILDREN
143
deposit - hyaline casts, occasional red blood cells, pus cells 2 or 3 per
H.P.F. and a very occasional granular cast.
Post-mortem examination revealed a well-nourished child with a
haemorrhagic infarct in the upper pole of the right kidney with an area of
softening surrounding it. The cerebral venous sinuses showed no evidence
of thrombosis. There was some thrombosis of the branches of the Sylvian
vein and the vessels over the left parietal region. The sulci in this region
showed thrombosis of the vessels on the suface of the cortex with slight
subarachnoid haemorrhage and a haemorrhagic necrosis in the surrounding
brain tissue. The other vessels of the brain were engorged, but no other
thromboses were found. There was pus in the right middle ear and mastoid
antrum which was cultured and grew a haemolytic streptococcus.
The above four cases have each shown an unusually high level of
chloride in the cerebro-spinal fluid. A summary of the cerebro-spinal fluid
findings is given in table 3.
Case 23. S. S., a boy, aged ten months, had been a normal breast-fed
infant until three months before admission, when he began to sleep for
lengthening periods until finally he could be aroused only with difficulty.
He developed a cough and bronchitis three weeks before admission to the
hospital. Examination revealed a well-developed and well-nourished pale
infant with irregular respirations. He was listless and drowsy, and disinterested in his surroundings. The eyes following a bright light and the
pupils reacted slowly. The discs were normal. There was a slight stiffness
of the neck. The limbs moved occasionally with stimulation but they
appeared weak, the legs being weaker than the arms. Muscle tone was
decreased and the reflexes were absent. Plantar response was extensor and
the Kernig sign negative. There was twitching of both arms and legs for
a short time after admission to the hospital. A lumbar puncture was
performed and clear cerebro-spinal fluid was withdrawn under no increase
of pressure. The cell count was 3 leucocytes per c.mm. The chemical
analysis revealed-sugar 25 mgm. per cent., chlorides 680 mgm. per cent. and
protein 13 mgm. per cent. In view of these findings further spinal punctures
were not performed. The fatty acid, cholesterol and phospholipoid estimations in the blood were within normal limits. The serum calcium was 9-8 mgm.
per cent. Four days after admission the child developed a temperature
of 1040 F. which subsided to 100° and 101° F., and was associated with a
right otitis media which did not appear severe. The child's mental condition
appeared to improve at times when he would take his feeds better and
occasionally be interested in his surroundings. A blood count showed:Haemoglobin 62 per cent., erythrocytes 3,500,000 per c.mm., leucocytes
3,800 per c.mm. Differential-Neutrophil polymorphonuclears, segmented,
12 per cent.; neutrophil polymorphonuclears, non-segmented, 12 per cent.;
eosinophil polymorphonuclears, segmented, 1 per cent.; lymphocytes, large,
46 per cent.; lymphocytes, small, 23 per cent.; lymphocytes, immature,
2 per cent.; monocytes 4 per cent. Two normoblasts were found in counting
100 leucocytes. There was some anisocytisis.
At post-mortem examination a terminal bronchopneumonia was found
in both lower lobes. On opening the skull the dura mater was found to
be thickened and adherent to the bone. The convolutions of the brain
appeared shrunken and the sulci very prominent. All of the superficial
vessels appeared to be distended and thrombosed. There was an excess of
cerebro-spinal fluid over the surface. All of the cranial venous sinuses
contained an old-standing thrombosis (fig. 7). The accessory nasal sinuses
were all found to be healthy, and also the left middle ear and mastoid
a144
ARCHIVES OF DISEASSE IN CHILDHOOD
antrum. The right middle ear and mastoid antrum and cells contained thick
pus, which grew a haemolytic streptococcus. Cultures of the thrombus
were sterile.
The possibilitv of some previous meningeal irritation which would
account for the lethargy and drowsiness for three months could not be ruled
out in this case. The more reasonable explanation, however, would appear
to be a partial thrombosis from the onset, possibly due to a respiratory
infection which was not recognized bv the parents.
FIG. 7. Case 23. S. S.-Thrombus lying in exposed superior longitudinal sinus.
Case 24. I. H., a boy, aged nine months, suffered with post-basal
meningitis for about six weeks before his admission to the hospital. The
head became hydrocephalic with some retraction. The cerebro-spinal fluid
contained 330 cells per c.mm., sugar nil, chlorides 617 mgm. per cent. and
protein 127 mgm. per cent.
At post-mortem examination a chronic basal meningitis with marked
internal hydrocephalus was found. The left lateral sinus was partially
thrombosed as a result of an acute mastoid infection on that side.
This case illustrates the changes which are found in the cerebro-spinal
fluid due to the original disease, in this case meningitis. Here the
thrombosis of the lateral sinus was of recent origin and probably resulted
from an acute mastoid infection. There were not any clinical signs which
would suggest the presence of sinus thrombosis.
Case 25. I. C., a boy, aged seven months, was a normal infant who
became sleepless and developed uncontrolled movements of the legs and
tongue. He was found to have an external strabismus, slight neck rigidity
and a positive Kernig sign on each side. The limbs were hypertonic and the
reflexes brisk. Chvostek's and Trousseau's signs were negative. The child
developed a convulsion in which the limbs were extended, the face was
distorted and there was marked cyanosis. Later the right arm and shoulder
was noted to be twitching. A lumbar puncture was performed daily for
three days, the first one showed only a few red blood cells. The third specimen contained 12 leucocytes per c.mm., sugar 54 mgm. per cent., chlorides
696 mgm. per cent. and protein 40 mgm. per cent. The child's condition was
variable, at times approaching normal, but usually extremely restless.
Finally the head became retracted, vomiting forceful, both legs rigid, the
head bathed in perspiration and the child screamed continuously unless
under the influence of sedative. He finally died four weeks after his
admission to the hospital and five weeks after the onset.
CEREBRAL SINUS THROMBOSIS IN CHILDREN
the veins draining into the superior logitudinal sinus.
FIG. 9. Case 25. I. C.-Thrombus in exposed
lateral sinus.
145
ARCHIVES OF DISEASE IN CHILDHOOD
146
At post-mortem examination the vessels over the surface of the
brain showed an extensive thrombosis. All of the cerebral venous sinuses
showed some thrombosis, which extended along the lateral sinuses and into
the right jugular foramen (fig. 8 and 9). Both middle ears contained a
moderate amount of thin purulent material. Cultures of this material grew
an haemolytic streptococcus and haemolytic staphylococcus albus.
Case 27. P. J., a girl, aged eight months, was a well-developed and
well-nourished infant, perfectly well until one week before being sent to
hospital, when she became drowsy and vomited occasionally. Later she
became very restless and pallor was noticed. The legs twitched occasionally
and finallv became stiff. Examination revealed a positive Kernig sign on
both sides and the neck was rigid. The knee jerks were present and equal.
Pus cells and albumin were found in the urine. The cerebro-spinal fluid
contained 35 leucocytes per c.mm. and 62 red blood cells per c.mm. A
differential count of the leucocytes showed 60 per cent. polymorphonuclears
and 40 per ceint. mononuclears. Cultures were sterile. A blood count
revealed a hypochromic anaemia: -Haemoglobin 45 per cent., red blood
cells 3,520,000 per c.mm., colour index 0 64 and leucocytes 19,000 per c.mm.
After showing an irregular, septic temperature for four days the child died
with generalized convulsions and cyanosis. At post-mortem examination
the superior longitudinal sinus and the cerebral vessels draining into it
showed an extensive thrombosis (fi-. 10). There was an area of sub-
FIG. 10.
Case 27. P. J.-Thrombus exposed by opening the superior
longitudinal sinus.
arachnoid haemorrhage over the upper part of the left Rolandic area. Both
middle ears and the mastoid antra contained thick creamy pus. The kidneys
showed several pyaemic abscesses. Cultures of material from the ears grew
H. influenzae and pus from the kidney grew B. coli.
Symptomatology
A studv of the eight cases which have been summarized reveals a number
of common signs which have assisted in arriving at a proper diagnosis. The
usual history obtained is one of a previously healthy child who suddenly
becomes drowsy or tired with periods of extreme restlessness followed by
convulsions or localized twitchings of some part of the body. The story then
goes on to paresis of limbs, face, tongue or eye muscles and sometimes
marked spasticity. Rhythmic movements and clonic spasms were frequent
and vomiting and feverishness were complained of in several instances. A
stiff neck, with or without head retraction, a positive Kernig sign and
absent reflexes were usual findings. A marked purpuric rash developed,
CEREBRAL SINUS THROMBOSIS IN CHILDREN
147
chiefly over the skin of the extremities in three children. A not uncommon
observation was irregular breathing and irregular pulse. The variability of
the mental state and periods of semi-consciousness or unconsciousness were
sometimes outstanding features. The temperature was extremely variable.
Thus in case 19 the temperature remained at 1020 F. throughout the illness,
whilst in case 20 it remained normal or subnormal except on two readings
when it reached 1000 F. The former showed far less evidence of infection
post mortem than the latter. Case 27 ran a swinging, high temperature
throughout, which was explained by the findings at autopsy of pyogenic
abscesses in the kidneys and gross infection of the mastoids and sinuses.
All patients developed some elevation of temperature during the course of
the illness, but this was often intermittent.
The above signs and symptoms are by no means confined to cerebral
sinus thrombosis, nor do they occur in every patient. However, they
occurred sufficiently often in this small series of cases that the condition was
usually considered in the differential diagnosis. Those cases associated with
direct extension from a mastoiditis to the lateral sinus, or of infection of
the face into the cavernous sinus, usually present a group of symptoms
which are more easily recognized and which are described in standard textbooks. These sinuses are not connected directly with the vessels which
drain the brain, and therefore if they become blocked there are alternative
routes for the return flow of blood, and severe cerebral symptoms do not
usually occur. Thrombosis associated with meningitis is usually marked
by the primary disease. The other cases form a group which show the foregoing signs and symptoms sufficiently often to produce a syndrome which
should be considered in the differential diagnosis of cerebral conditions in
children.
Differential diagnosis
The other cerebral conditions which give rise to signs and symptoms
resembling sinus thrombosis are meningitis, cerebral abscess, cerebral
haemorrhage, cerebral tumour and encephalitis. The most difficult of these
to rule out in the diagnosis of cerebral sinus thrombosis is often cerebral
haemorrhage, since there is some haemorrhage usually present with
thrombosis. However, the amount of haemorrhage and therefore the
number of red blood cells per c.mm. is usually much less in sinus thrombosis
than in cerebral haemorrhage. The signs and symptoms are more constant
with haemorrhage and usually more sudden in onset. Purulent meningitis
is usually easily differentiated from sinus thrombosis, but tuberculous and
post-basal meningitis present some similar findings. In purulent meningitis
the fluid contains large numbers of polymorphonuclear cells. In all forms
of meningitis the sugar content is reduced or absent; the chlorides are
lowered, this being most marked in tuberculous meningitis; the protein
is higher than in sinus thrombosis. Cerebral abscess and cerebral tumour
lead to papilloedema much more frequently than thrombosis, and their
onset is much slower. The clinical picture of sirnus thrombosis is much more
ARCHIVES OF DISEASE IN CHILDHOOD
148
acute than the gradual procession of signs and symptoms in abscess and
tumour. The cerebro-spinal fluid is usually unaltered in the two latter
conditions. Occasionally the cell count may be slightly increased in brain
abscess, but here the presence of red blood cells, xanthochromic fluid, or
increased chlorides may help in making a diagnosis of sinus thrombosis. The
absence of polymorphonuclear leucocytes in the cerebro-spinal fluid in
encephalitis is a differential point, but the signs and symptoms are often
similar to sinus thrombosis. Experimental work recently reported by
Frenckner' suggests a possible diagnostic aid by the use of x-rays and the
injection of x-ray opaque substance into the superior longitudinal sinus.
This method shows the point at which the thrombus causes obstruction, and
in some cases might lead to successful operative interference. At present
the prognosis is hopeless in the majority of cases unless the thrombosis is
confined to the lateral sinus.
The cerebro-spinal fluid
Patients suffering from thrombosis of cerebral vessels usually present
signs and symptoms which indicate a severe intracranial lesion and therefore
a lumbar puncture becomes almost essential in making an accurate
differential diagnosis. Reference to table 3 brings out several points in the
TABLE 3.
LEUCOCYTES RED BLOOD CELLS
per c mmn.
per c.mm.
SUGAR
CHLORIDES
PROTEIN
mgm. per cent. mgm. per cent. mgm. per cent.
Case 19 (1)
6
1,000
56
876
90
(2)
8
Xanthochromic
-
932
125
Case 20
4
0
0
774
13
Case 21
6
62
48
766
23
Case 22
8
17
70
866
10
Case 23
3
0
2a
680
13
Case 24
330
0
0
617
127
Case 25 (1)
4
90
-
(2)
6
Slight blood tinge
-
(3)
20
1,950
54
696
40
35
62
-
-
Case 27
examination of the cerebro-spinal fluid which are of considerable importance
in arriving at a diagnosis of cerebral thrombosis. The leucocytes are usually
within normal limits but occasionally may be slightly raised. (In case 24
the increase to 330 per c.mm. was a result of the meningitis.) A differential
CEREBRAL SINUS THROMBOSIS IN CHILDREN
149
count of the leucocytes usually shows a preponderance of polymorphonuclears. Red blood cells may, or may not be present. The number of these
varied from 17 to 1,950 per c.mm. in five out of the eight cases in which
the fluid was examined. The presence of red blood cells, excluding
traumatic haemorrhage of course, is practically always due to some degree
of subarachnoid haemorrhage in the region of the thrombosed vessels. This
was true in all of the cases in table 3 which showed red blood cells in the
cerebro-spinal fluid, while the others with an absence of red blood cells
showed no evidence of subarachnoid haemorrhage.
Greenfield and Carmichael'" state that every case of cerebral thrombosis
is associated with some haemorrhage from capillaries and that this accounts
for the few red blood cells and the not uncommon finding of xanthochromic
fluid. Ehlers and Courville' emphasize the importance of a bloody or
xanthochromic fluid in the diagnosis.
The chemical changes in the cerebro-spinal fluid were particularly
interesting. The amount of sugar present varied from normal to a complete
absence and in the presence of meningitis the sugar was lowered or
completely absent. Unless complicated by meningitis the cerebro-spinal
fluid proteins were within normal limits or slightly raised, the values tending
to be higher in those specimens which contained red blood cells. There is
usually no record of the chloride content of the spinal fluid in the cases which
have been reported in the literature. It will be seen that the majority of
the cases in this series in which estimations were made on the cere'bro-spinal
fluid the chlorides reached a high level. This occurred sufficiently often to
make it almost diagnositic. In four cases the high value for chloride
suggested the possibility of cerebral sinus thrombosis. Figures of 876, 932
and 866 mgm. per cent. are strikingly high and have not been encountered
in other conditions in children. An analysis of one thousand cerebro-spinal
fluid chloride estimations carried out in this hospital reveals no figures as
high as these. Neale and Esslemont" working in this hospital found the
chloride values in non-meningitic cases varied from 696 mgm. per cent. to
783 mgm. per cent. with average of 730 mgm. per cent. Greenfield and
Carmichael" give the normal range of cerebro-spinal fluid chlorides as 720
to 740 mgm. per cent., and state that a chloride value above 750 indicates
renal deficiency with impermeability to chlorides. A histological study of
the kidneys in these cases of cerebral sinus thrombosis has not revealed any
changes which might explain an impermeability to chlorides. Whilst Neale
and Esslemont1' showed that the amount of chloride in the cerebro-spinal
fluid varies to some degree with that in the blood, the administration of
sodium chloride solutions intravenously, or by other routes, will not explain
the high figures, since only two of the patients received rectal saline.
Blacklock and Morris"3 have recently reported interesting work on experimenltal chloride retention in rabbits. They have shown that there is a
retention of chloride in heart, liver, brain and skeletal muscle in the presence
of acute infection. The possibility of this retention being due to the kidneys
becoming less permeable to chloride is, in their opinion, difficult to believe
150
ARCHIVES OF DISEASE IN CHILDHOOD
because the kidney has not lost its ability to concentrate and excrete urea
and other substances. This seems to be borne out in case 21, whose blood
fortunately was examined at the same time as the cerebro-spinal fluid. The
cerebro-spinal fluid chlorides were 766 mgm. per cent., while the blood urea
was 23 mgm. per cent. and the non-protein nitrogen 21 mgm. per cent.
Causation of thrombosis
Byers and Hass5 have reviewed many possible etiological factors in the
mechanism of thrombosis, such as the severe loss of fluids in diarrhoea and
vomiting and the associated polycythemia in some cases. The administration of blood and other fluids has been suspected as a possible cause of
thrombosis, but the above authors found that not all of their patients had
received such treatment, nor could they find evidence of destruction of the
endothelial lining by bacteria or toxins. Simpson7 felt that intravenous
therapy via the anterior fontanelle into the superior longitudinal sinus played
a large part in the formation of thrombosis in some of his cases. Case 17
in this series received a blood transfusion by this route, but the others had
not received any intravenous therapy which could be of etiological
significance. Ashchoffl0 speaks of the relation of the blood flow to
the formation of thrombi, and slowing of the blood from various causes is
favoured by Byers and Hass5 as the most important etiological cause.
Although there was not an opportunity to carry out any special studies
in the present series of cases, in view of the above theories several facts seem
to be of importance. In the first place, none of the children had suffered
any appreciable loss of body fluids before the signs of thrombosis were well
established. None was associated with diarrhoea although in a few instances
occasional vomiting occurred in the early stages of the illness. Intravenous
therapy had not been used in any of the cases and thrombosis has never been
found at post-mortem examination on several hundred patients who have
received intravenous therapy. The extremely sudden onset of symptoms
in the majority of these cases pointed to the early establishment of a
thrombus and does not fit in with any theory of dehydration or stagnation
of blood in the sinuses. Every patient showed evidence of some septic
process at post mortem, even though it was not demonstrable during life,
and this appears to be the only common factor. Why such infections should
produce changes in sinus wall, which cannot be demonstrated, but which
give rise to thrombosis, is difficult to explain, especially when it is considered
that in a high proportion of children infection in some part of the body and
usually in the middle ears is found at autopsy. It is recognized that convulsions are not uncommon during this age period in association with infections
sometimes of a trivial nature; and it is conceivable that convulsions might
have preceded the thrombosis in some of these cases, possibly producing
conditions favourable to thrombus formation in the sinuses. Thrombosis is
occasionally attributed to anaemia. An anaemia was present in only three
of the cases recorded here, the haemoglobin values being 45, 62 and 29
per cent., and its importanee is difficult to assess in comparison to the
presence of infection.
CEREBRAL SINUS THROMBOSIS IN CHILDREN
151
A study of table 4 shows the possible etiological factors in this series
of cases. Case 12 was the only one which did not have a demonstrable focus
of infection, the only known factor here being a moderately severe
nutritional anaemia.
TABLE 4.
POSSIBL.E ETIOLOGICAL FACTORS IN 32 CASES.
PRIMARY EAR INFECTION
...
EAR INFECTION FOUND P.M. ...
ACCESSORY NASAL SINUS INFECTION
PURULENT INFECTION ELSEWHERE
PNEUMONIA
...
...
ANAEMIA
TERMINAL DIARRHOEA
...
PULMONARY TUBERCULOSIS ...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
3 ...
...
...
...
...
...
...
...
9
9
6
7
7
2
1
The conclusion to be drawn from these cases is that infection is probably
the chief factor in t-he production of thrombosis, since, although it may not
be detectable during life, it is nevertheless practically always found at
autopsy. Toxins may be conveyed by the blood stream, producing an effect
upon the vessels which leads to thrombosis.
Summary
Cerebral sinus thrombosis, although comparatively rare, appears to be
increasing. Eight case summaries are reported with important facts
relating to thirty-two cases recorded during fourteen years. A review of
the principle signs and symptoms in these cases suggests that there is a fairly
constant clinical syndrome associated with the majority of those cases which
are not due to mastoiditis and cellulitis with involvement of the lateral and
cavernous sinuses, respectively. There are in this group the following points
which often assist in making a correct clinical diagnosis of sinus
thrombosis: -Sudden onset of restlessness followed by convulsions; localization of the convulsions as shown by twitching and followed by paresis; cerebrospinal fluid findings of increased or normal pressure, cells normal or slightly
increased, the presence of red blood cells or xanthochromic fluid, cultures
sterile, low sugar content, high chlorides and normal or slightly raised
protein. The other points in the condition are the drowsiness and semiconsciousness; the rapid decline of general health in a previously healthy
child, the post-mortem findings of thrombosis of cerebral vessels and in
some cases the absence of any evident infection during the course of the
illness. There is practically always an infection found at post mortem.
The presence of high chloride values in the cerebro-spinal fluid is emphasized
as an aid in arriving at the diagnosis. Several theories have been reviewed
which attempt to explain the cause of the thrombosis, but no reference hlas
been made to treatment or detailed pathology.
Thanks are due to the Honorary Physicians of the Children's Hospital
for permission to use their case records,
152
ARCHIVES OF DISEASE IN CHILDHOOD
REFERENCES
1. Morgagni, J. B., quoted by Byers, R. K., & Hass, E. M., Am. J. Dis. Child.,
Chicago, 1933, XLV, 1161.
2. Abercrombie, John, quoted by Byers, R. K., & Hass, E. M., loc. cit.
3. Ehlers, H., & Courville, C. B., J. Pediat., St. Louis, 1936, VIII, 600.
4. Holt, L. E., Jnr., & McIntosh, R., Holt's Diseases of Infancy and Childhood,
tenth edition, London, 1933.
5. Byers, R. K., & Hass, E. M., Am. J. Dis. Child., Chicago, 1933, XLV, 1161.
6. Wyllie, W. G., Brit. J. Child. Dis., London, 1924, XXI, 124.
7. Simpson, R., Canad. Med. Ass. J., Montreal, 1932, XXVI, 317.
8. Osler, W., Principles and Practice of Medicine, tenth edition, London, 1927.
9. Frenckner, P., Proc. Roy. Soc. Med., London, 1937, XXX, 413.
10. Aschoff, L., Arch. Int. Med., Chicago, 1913, XII, 503.
11. Greenfield, J. E., & Carmichael, E. A., Cerebro-spinal Fluid in Clinical
Diagnosis, London, 1925.
12. Neale, A. V., & Esslemont, M. S., Arch. Dis. Childh., London, 1928, III, 243.
13. Blacklock, J. W. S., & Morris, N., Jour. Path. c8f Bact., Edinburgh, 1936, XLII,
329.